I think it's quite reasonable to say that a language such as Burushaski is not coded in ISO 639-2 and that mis is a special-purpose ID that can be used to handle content in such a scenario, just as Cobol is not coded in ISO 639-2 (for different reasons) and zxx is a special-purpose ID that can be used to handle such content.

If we are to say that Burushaski *is* coded, then we have a formal problem that nothing whatsoever in ISO 639-2 clearly documents that denotation, as well as a problem that the denotation would change if ever Burshaski was (directly) coded. These problems are avoided if we cast things as above: it's not coded, and mis is an ID for that general scenario.

Note that tagging content with mis does *not* at all tell you the content is Burushaski -- applying ISO 639-2. (You might argue that the interpretation would be different under MARC, but that's MARC, an application of ISO 639, not ISO 639 itself.) What this *does* tell you, as the intended ISO 639 semantic, is: "The content is in some human language variety, but it's not encompassed by any (otherwise) coded language variety." To me, it's a fair stretch to say that that semantic entails or implies Burushaski in any direct way. Note that an application of ISO 639-2 could very well code Burushaski as qaa -- in which case the semantic clearly should not entail or imply Burushaski.

The crux of the issue is whether mis is a collective ID or a special-case ID. If it's a collection, then we could say that a language like Burushaski is coded in ISO 639-2; but we have several problems, some quite serious (including that in a given application Burushaski might be coded with a private-use ID, creating duplicate coding). If it's a special-case ID, we would not say Burushaski is coded, and most problems are avoided. (The one problem we don't avoid that content tagged mis is at risk if the cases covered by mis in a given application context are subject to change. That's a completely unavoidable problem, however.)


Peter


-----Original Message-----
From: ISO 639 Joint Advisory Committee [mailto:[log in to unmask]] On Behalf Of Milicent K Wewerka
Sent: Friday, May 04, 2007 7:45 AM
To: [log in to unmask]
Subject: Re: decisions required: "other" collections, mis

But since "mis" is in the code list, how can the languages be "uncoded"?  This symbol is simply used for languages that don't fit into any of the other available collective groups.

Milicent Wewerka

>>> Debbie Garside <[log in to unmask]> 05/04/07 10:26 AM >>>
"Uncoded languages" would be my preference.

Best regards

Debbie

> -----Original Message-----
> From: ISO 639 Joint Advisory Committee
> [mailto:[log in to unmask]] On Behalf Of Peter Constable
> Sent: 04 May 2007 15:03
> To: [log in to unmask]
> Subject: Re: decisions required: "other" collections, mis
>
> From: ISO 639 Joint Advisory Committee
> [mailto:[log in to unmask]] On Behalf Of Rebecca S. Guenther
>
> > I would agree with the intent...
>
> Good. Thanks for confirming that.
>
>
> > Also, in the sentence below...
> > it would be clearer to say:
> > "If a new language is added to ISO 639-2 which was
> previously listed
> > as a language under "mis"...
>
> A problem with that is that ISO 639 has never listed
> languages under mis.
>
>
> > I would prefer calling it something like "Other languages"
> > or "Other unrelated languages". Saying "Unsupported languages"
> > doesn't make sense to me-- it's not clear what isn't supported.
> > If they're really "unsupported" there wouldn't be an identifier for
> > them. It's really more "Unenumerated languages"-- or miscellaneous
> > languages that don't belong in any defined group.
>
> Joan indicated "unsupported" was better to her than
> "miscellaneous". I see what you say about "unsupported", though.
>
> Some possibilities:
>
> - Unsupported languages
> - Other languages
> - Other unrelated languages
> - Unenumerated languages
> - Uncoded languages
> - Other uncoded languages
>
> Or maybe others have other ideas.
>
> Perhaps it might be useful if each of us indicated a couple
> of choices in order of preference. My picks:
>
> 1) Other languages
> 2) Uncoded languages
>
>
>
> Peter
>
>
>